ai icon indicating copy to clipboard operation
ai copied to clipboard

Text to Speech utils?

Open nabilfatih opened this issue 2 years ago • 4 comments

Feature Description

Love to see how AI SDK can handle Text to Speech from OpenAI. As I see from documentation, TTS can be streamed. https://platform.openai.com/docs/guides/text-to-speech/streaming-real-time-audio

Use Case

Chatbot but with speech. Like the chatGPT application on mobile.

Additional context

No response

nabilfatih avatar Dec 26 '23 12:12 nabilfatih

need an intergration with google cloud apis or an server pre-hosted

01582 avatar Dec 29 '23 08:12 01582

you could do that with https://replicate.com/ and some open source project for text-to-speech

01582 avatar Dec 29 '23 08:12 01582

i don't think that vercel would add the text-to-speech utils

01582 avatar Dec 29 '23 08:12 01582

Yes we will have: https://github.com/vercel/ai/pull/922

Now , @lgrammel I wonder if we could also add some STT to the text input to make it end-to-end conversational.

I'm experimenting with https://github.com/JamesBrill/react-speech-recognition

llermaly avatar Jan 27 '24 16:01 llermaly